Assessing event-based college student drinking and social context using mobile devices

nasa-89127-unsplash.jpg

THE CHALLENGE

Most surveys of alcohol-use ask about behaviors and patterns that happen days, weeks months or even years ago. It's all retrospective recollection from study participants. It's useful for identifying general trends but it's fraught with measurement errors.  And when the topic of the study is binge drinking, errors involving memory can increase. -- even if you're asking the question the next day.

 

SOUNDROCKET USES INNOVATIVE TECH TO SOLVE RESEARCH PROBLEMS

We’re working with a team at the University of Minnesota School of Public Health to identify better ways to capture data in the moment, when it is most likely to be accurate and least impacted by recall errors.  Mobile web-based surveys coupled with SMS (text) messaging are providing a new tool for doing this. We capture survey data in the moment with very short questionnaires that are implemented very quickly via SMS invitation. This allows us to get to participants when the experiences are fresh and unfolding-- sometimes only minutes or hours after events take place, rather than days, weeks, etc.

In one study conducted during a homecoming weekend we surveyed over 1,700 students. Students were asked to participate in a sequence of eight surveys over the course of a series of weekend Homecoming events (including a Friday night concert, a Saturday football game, and Saturday night parties).

WHY WHAT WE DID IS UNIQUE

Short surveys. Mobile optimization.  Invitations to participate via SMS.  On their own, all three technologies prevalent in today’s world – yet rarely integrated into a data collection protocol as we did.  Most web based survey invitations are sent by email -- but having participants mobile numbers with approval (obtained by the participants) to use them for the study, we rapidly sped up the process of participation.

Over 88% of those invited participated in the study! And generally, most participants did so within minutes of being invited.  This quick response allowed us to time specific questionnaires to match the events happening on campus.  For example, separate pre- and post-concert surveys were conducted.  Similarly, pre- and post-football game surveys were administered.

The researcher team used the data to show the importance of the role of social context and expectations about alcohol use influences negative consequences around college student drinking.

Our collaborative success with the University of Minnesota research team on this project has launched SoundRocket towards doing exploring other innovative opportunities to further integrate mobile technologies into situational or event based research.